PROJECT SUMMARY (PROJECT 2, Project Leader: Dougherty, Massachusetts General Hospital) The goal of Project 2 is to causally probe the role of cortico-striatal circuits in approach-avoidance conflict using invasive recording and neurostimulation in awake, behaving humans. Invasive studies in humans serve as a bridge between the complementary invasive animal and non-invasive human studies in other Projects. Noninvasive human functional imaging studies provide valuable insights into the circuitry underlying specific brain functions but ? fundamentally ? are correlative in nature. Invasive measurements in humans during tasks provide markedly better spatial resolution (down to the single neuron) and temporal resolution (down to 5-10 milliseconds). More critically, the ability to stimulate using the same electrodes allows for unique ?backwards neuroimaging?. Instead of performing a task and observing which brain networks are involved, one can stimulate different brain regions at different amplitudes, frequencies and task epochs to observe effects on task performance all while simultaneously recording. In this project we propose to study approach-avoidance conflict in two human models: (1) deep brain stimulation (DBS) of the ventral capsule/ventral striatum (VC/VS) and (2) surgical epilepsy monitoring. Over five years, we will recruit 20 DBS patients with chronic DBS electrodes in the VC/VS for treatment of neuropsychiatric disorders and 20 patients with epilepsy with acute depth electrodes for seizure monitoring. We will investigate the behavioral and neural effects of DBS on approach-avoidance behavior in patients with major depressive disorder (MDD) with implanted DBS electrodes in the VC/VS. Capitalizing on our prior studies and extensive set of preliminary data we will use an approach/avoidance decision making task to assess behavioral and neural changes associated with stimulation of the ventral striatum (as opposed to no stimulation) in neuropsychiatric patients with implanted DBS depth electrodes. We will also investigate the large- scale neural network of approach-avoidance behavior in epilepsy patients implanted with depth electrodes, and test the hypothesis that closed-loop stimulation targeting key cortico-striatal nodes will bidirectionally modulate approach-avoidance behaviors in depressed patients. Building on our extensive experience in this area we will record simultaneous activity of multiple prefrontal structures and, in some cases, striatum, in patients admitted to the MGH Epilepsy Monitoring Units following implantation of depth electrodes. To maximize synergies across Projects, the task will be identical to the one used in Project 1 (with unmedicated individuals with MDD or anxiety disorders) and functionally analogous to the one in Project 3 (with non-human primates). Contribution to Overall Center Goals & Interactions with Other Center Components. By testing that direct stimulation of both approach-related (e.g., striatum) and avoidance-related (e.g., pregenual anterior cingulate cortex) areas affects approach-avoidance behavior in humans, Project 2 will directly link the non-invasive fMRI studies in individuals with unmedicated MDD (Project 1) with the non-human primate (Project 3) and rodent (Project 4) studies that will mechanistically dissect neurobiological mechanisms governing approach/avoidance behaviors.